Discussion of “Axial Compression of Footings in Cohesionless Soils. I: Load-Settlement Behavior” by Sami O. Akbas and Fred H. Kulhawy

Abstract

The discussers would like to commend the authors for their investigation of the characterization and parameterization of the load-settlement behavior of axially loaded footings on cohesionless soils because of its importance in geotechnical practice. In this discussion, the discussers would like to: 1 provide the results of a comparative assessment of the performance of loaddisplacement models in approximating the results of a set of fullscale load tests on large footings; 2 stress the importance of investigating load-displacement uncertainty by addressing quantitatively not only second-moment statistics of model parameters but also the correlation between the parameters themselves; and 3 examine the adequateness of load-displacement models for the generation of multivariate distributions for simulation purposes. A database of 30 footings on 12 different sands that were subjected to vertical loading in the field was used for these purposes. The origins of these sands included natural marine deposits, alluvial sediments, aeolian dune sands, and compacted sand fills. The footings included 18 square, 7 rectangular, and 5 circular shallow foundations, with the equivalent footing width B varying from 0.5 to 6 m. Only full-scale foundations were considered to avert difficulties associated with 1-g model tests and scaling problems e.g., Cerato and Lutenegger 2007 . A summary of the sand sites, test locations, sand origin, index number in the database, footing sizes, embedment depth, groundwater depth, median grain size, and CPTU net cone tip resistance qnet representative of capacity is given in Uzielli and Mayne 2010 . The applied stress was normalized by qnet; settlement s was normalized by foundation width B. Three load-displacement model types were considered: